Agreement Between Methods to Estimate Residual Lung Volume: A Methodological Investigation

Abstract

Introduction: The determination of residual lung volume are technically challenging and can involve rather elaborate techniques. However, due to the complex nature of measurement protocols, a number of studies have attempted to use alternative estimation techniques, including application of regression equations following spirometry measurement, panting manoeuvres and general predictive equations. With such extensive measures, it is difficult to reach a consensus where all residual lung volume measures are in agreement, hence the aim of this methodological investigation.
Methods: Twenty two participants (n=10 male and n=12 female) were recruited from the University of Gloucestershire, undergraduate programmes. All participants were over 18 years of age and all were free from disease, illness or injury (͞χ±s; age=20.5±1.7 years, body mass=68.7±1.5 kg and stretched stature=172.0±8.3 cm). Three estimations of residual lung volume were carried out by participants, a ‘spirometry’ method (via forced vital capacity), a ‘panting’ method (via the air displacement plethysmograph (BOD POD®)) and a general ‘prediction’ method (based on age, gender, ethnicity and stature predictive equation to estimate whole body density (Db)). Data analysis was conducted to establish the linear relationship and agreement between the three estimation methods by constructing scatter plots showing deviation from the line of identity and by applying the 95% limits of agreement (LoA) method to quantify the bias, random variation and heteroscedasticity.
Results: Results indicated that linear relationships were evident from the scatter plots, but this was expected given they were measuring the same variable. Further analysis with limits of agreement indicated that there was a bias of 0.13, 0.17 and 0.04 l for the panting, spirometry and prediction estimation techniques and limits of agreement of 0.47 to -0.21, 0.45 to -0.11 and 0.23 to -0.15 L respectively.
Conclusion: The spirometry technique demonstrated a more accurate estimation of residual lung volume when compared to panting and prediction techniques, in addition, as spirometry uses standard (and the simplest) techniques to determine lung volumes, and is the most widely used method within research determining Db from hydrostatic weighing, it was concluded that the spirometry method would be the measurement approach of choice for determination of residual lung volume.